the aid policy network in pakistan: actor network...
TRANSCRIPT
The Aid Policy Network in Pakistan: Actor-Network Analysis
Faheem Jehangir Khan
Abstract
Despite an extensive debate on foreign aid, the voluminous literature mainly
focuses on the evaluation of foreign aid outcomes (or aid effectiveness). The
aid literature has rarely considered the aid policy process and the influence
of aid policy networks on managing foreign aid decisions in an aid recipient
country. Using the actor-network analysis, this study helps to understand the
complexities, interdependencies, and constraints involved in managing aid
policy process in Pakistan. The findings indicate that absence of an inclusive
actionable development framework and limited capacity of the state provide
opportunities for donors to dominate the policy debate and pursue their own
priorities. Further, the provincial devolution of 2011 has made the aid
delivery system in Pakistan even more fragmented and complex, operating at
multiple levels.
JEL Classification: F35, H50, H79, H83, O20
Key words: foreign aid, policy networks, network management, policy process, Pakistan
1
Over the last 65 years, more than $3 trillion has been spent on international development
programs in collaboration with development partners (Giugale, 2014). Extensive research
and analytical work has been carried out on foreign aid in the last six decades. However,
the contribution that foreign aid can make to development is still contested. The large and
growing aid literature mainly focuses on donors’ strategies, aid modalities, and the scope
for foreign aid to improve development indicators. The evaluation of foreign aid outcomes
(or aid effectiveness) dominates the aid debate. The aid literature has rarely considered the
practical working of the aid policy process – specifically, how aid decisions are managed
and transformed into action – and the influence of policy networks on managing decisions
related to foreign aid and development priorities in an aid recipient country such as that of
Pakistan. In that sense, apart from examining aid effectiveness, it is equally important to
understand the complex policy networks that influence the aid policy process. This
research responds to this gap in the literature by seeking to explore the network structure
in place to manage foreign aid in the complex aid policy network1 in Pakistan.
The aim of this research is to study the aid policy network in Pakistan.2 The objective of
this study is to explore donor-government interactions, including around the Millennium
Development Goals (MDGs), in managing the aid policy process in Pakistan. In this context,
this research attempts to map the complex aid policy network, determine resource
interdependencies, examine actors’ interaction patterns, and explore actors’ perceptions
about managing aid policy process in Pakistan. Adopting Koppenjan and Klijn (2004) and
Klijn and Koppenjan’s (2016) actor-network analysis, this research will help us to
understand the complexities, interdependencies, and constraints involved in managing
foreign aid in Pakistan.
Pakistan became a recipient of foreign aid in the 1950s and today it is among the top ten
recipients of official development assistance (OECD, 2015). Since the independence of
Pakistan from British rule in the Sub-continent (1947), foreign aid inflows have a strong
linkage with the geo-political interests of donors (Ali, 2009). In the 1950s the inflows were
1 The current study focuses on donor-government aid relationships within a network of multiple actors, therefore it corresponds to the aid policy network. 2 An initial version of this paper has been published in PIDE Working Paper Series, No.140:2016.
2
negligible, but the 1960s witnessed a rapid increase in aid inflow which was connected
with mutual economic and defence agreements between General Ayub Khan’s regime
(1958-1969) and the United States during the Cold War era. Aid inflows remained low
during the 1970s, but again had a sharp increase during the 1980s due to the Afghan-Soviet
war (1979-1989). Following the demise of the Soviet Union, aid to Pakistan was cut and
economic and military sanctions were imposed under the “Pressler Amendment” (to
reduce significantly the risk that Pakistan would possess a nuclear explosive device) in the
1990s at the time when Pakistan's cooperation was no longer needed. In 1993, additional
sanctions were imposed under the ‘Missile Technology Control Regime’ (MTCR) for
allegedly receiving missile technology from China and North Korea. Aid inflows to Pakistan
were further cut down after the 1998 nuclear tests and General Musharraf’s military
takeover in October 1999. A new era of high aid inflow was observed in the post 9/11
period that resulted in closer Pakistan-US ties over the ‘war on terror’ in Afghanistan.
Since 1951, the Pakistan government has received over $101 billion (in nominal terms)
under ODA from official bilateral and multilateral donors of which $41 billion was
disbursed during the last thirteen years (PBS, 2011 and EAD, 2002-2015), i.e., since the
inception of the MDGs. Aid as a percentage of Pakistan’s Gross National Income (GNI) has
mostly remained in the range of 0.9-1.7 percent since 2001. In absolute terms, aid per
capita fluctuated between $5-20 and total ODA between $730 million to $3.5 billion during
this period (GOP, 2013). There are several measures to evaluate aid trends and its
composition, but Bräutigam (2000) suggests that aid as a percentage of government
expenditure would probably be the most useful measure for a study of aid dependence and
governance. Figure 1 presents foreign aid to Pakistan as a percentage of government
expenditure during 2000 and 2015. The graph shows a declining trend of official
development assistance as a percentage of government expenditure during the early years
of MDGs, while a rising trend since 2013.
3
Figure 1: Official development assistance as % of government expenditure (2000-2015)
Source: Author’s illustration
This paper is structured as follows. Section 2 gives an overview of the existing aid
literature. Section 3 presents the research methodology and sets out the theoretical
framework. Section 4 presents findings and discussion focused on an actor-network
analysis of the aid policy network in Pakistan. Finally, section 5 sets out the conclusions of
this research.
LITERATURE REVIEW
Foreign aid has been one of the most contentious topics in the history of intellectual
thought on international development (Mavrotas, 2010). Extensive research and analytical
work has been carried out on foreign aid in the last six decades. Contributing to the aid
debate, authors have analysed almost every aspect of foreign aid that has considerably
improved our knowledge and understanding about its problems and effectiveness.
4
The aid delivery system is often perceived as if it was a linear chain that links a donor
government to a recipient country beneficiary via various intermediary organisations
(Gibson et al., 2005). This notion ignores the complex network of diverse and
interdependent actors involved in the aid delivery system. Over the last six decades, with
the steady evolution of global events, the aid delivery system has transformed
considerably. The numbers of official aid agencies, their partner organisations and budgets
have grown enormously. Kharas (2009) and Fengler and Kharas (2010) note that the aid
architecture has changed fundamentally since the aid landscape now includes new players3
that channel significant volumes of foreign assistance.
In the 1970s there were few bilateral and multilateral donor organisations. The OECD
(2011a) reports that 45 countries and 22 multilateral organisations disbursed ODA in
2011; the data did not include several increasingly important non-traditional donors, such
as the BRICS.4 Kharas (2009) notes that there were around 126 bilateral agencies from
DAC-member countries and 23 bilateral agencies from non-DAC member countries since
many countries have several agencies providing ODA. In another count, some 263
multilateral aid agencies give money to promote development while 56 countries provide
bilateral foreign assistance through several agencies (Fengler and Kharas, 2010).
Describing the situation, Moyo (2010) notes that some 500,000 people are directly
involved in the aid delivery system today.5
In the case of Pakistan, the aid establishment is now substantial. In 2013, there were 32
official bilateral and 12 official multilateral donor agencies (plus 18 UN agencies) operating
in Pakistan, managing some 2094 project activities in collaboration with government
organisations and local partners (OECD, 2015). At the recipient end (the government), 38
divisions of 33 federal ministries and around 150 provincial departments (38 Punjab; 36
Sindh; 32 Khyber Pakhtunkhwa; and 39 Balochistan) and executing agencies in four
3 Such as private aid organisations (foundations and private corporations), new bilateral donor agencies, thousands of privately funded international NGOs, and perhaps hundreds of thousands of community-based and civil society organisations in developing countries. 4 Brazil, Russia, India, China and South Africa 5 The World Bank employs 10,000 people; the IMF over 2,500; around 5,000 for UN agencies; add to that the employees of at least 25,000 registered NGOs, private charities and the army of government aid agencies
5
provincial governments interact with these 44 official donor organisations to manage
foreign aid in Pakistan. Further, as of 2006, some rough estimates (from UNDP and the
Agha Khan Foundation, Pakistan) suggest that there were around 100 international non-
government organisations (INGOs) and some 15,000 to 20,000 local non-government
organisations/civil society organisations (NGO/CSOs) operating in the country. The
presence of these numerous actors (or organisations) in the country, their contrasting
priorities for the development process, and overlapping project activities add up to the
complexity of aid delivery in Pakistan. With this transformation, the current aid
architecture no longer resembled its initial design, the foundations of which were laid with
the creation of the Bretton Woods Institutions and the success of the Marshall Plan
(Kharas, 2009).
Exploring the complex linkages between actors in the aid delivery system, Gibson et al.
(2005) identify multiple relationships between aid actors and examine the role of
institutional incentives in India, Sweden and Zambia. Describing the complexity of aid
relationships, Gibson et al. (2005) developed a schema they called the ‘international
development cooperation Octangle’.6 Rather than linear linkages among actors in the aid
delivery chain, the authors present a more realistic “tangle” of relationships involved in the
aid delivery system. Similarly, focusing on aid dependence and governance, Bräutigam
(2000) identifies three sets of players within the development aid process: bureaucrats and
politicians (within the government); interest groups and the electorate (those in the
society); and aid donors. Examining government actions in a game setting, Bräutigam
(2000) argues that the choices of these groups of actors are a function of their interests and
ideas, the incentives they face, and the formal and informal rules that shape the
possibilities for action.
The literature on policy network and public policy process denote that modern decision-
making process is highly complex. Every initiator of a decision-making process is
6 They identified eight major actors within the aid system which include: the donor government; the recipient government; other donors; the donor’s international development agency; sectoral ministries and agencies within the recipient government; third-party implementation organisations, including NGOs, private consultants and contractors; organised interest groups and civil society organisations within the donor and recipient countries; and, the target beneficiaries.
6
dependent upon a wide variety of other network actors, and their resources, to achieve
meaningful outcomes (Klijn, 2007). Similarly, today’s aid policy network is far more
complex, involves a range of actors, structures and processes. Since there is a need to
understand this constantly changing process that shapes foreign aid decisions, this study
seeks deeper understanding of the network structure and processes shaping foreign aid
decisions in Pakistan.
METHODOLOGY
A single-method approach to collect data was selected in this research, namely, semi-
structured interviews. Using a purposive-snowball sampling technique, qualitative
interviews were conducted with a range of actors from the aid policy network in Pakistan.
The data were then transcribed, coded and analysed, and results were interpreted. Ritchie
and Lewis’s (2003) Framework (or thematic framework) method was used to analyse the
data.
Reflecting the exploratory nature of the research, semi-structured interviews were selected
as the most appropriate research method for gathering primary data. A one-to-one
‘executive interview’ approach was adopted to collect the data. It was anticipated that
interaction in person with the participants – executives and senior officials – would allow
the researcher to involve the respondent in discussion and record detailed information.
Piloting was carried out prior to the fieldwork and audio recording was recommended
during the interviews with the consent of the respondents.
This study adopted the purposive sampling technique to extract the sample. This sampling
strategy is preferred when a specific (limited) number of individuals carry the most
relevant information that is sought (Bryman, 2012). Groups of actors were identified in the
aid policy network in Pakistan, followed by identification of actors and selection of cases in
the relevant groups. Further, there was also a degree of snowballing. Incorporating a
snowballing approach, we can say that this research actually used the purposive-snowball
sampling method.
7
Table 1: Selection of cases
APPROACHED POTENTIAL
RESPONDENTS USING
PURPOSIVE SAMPLING
NON- RESPONSE/
INCOMPLETE
APPROACHED ADDITIONAL
RESPONDENTS USING
SNOWBALL SAMPLING
TOTAL
DONORS 10 -3 5 12
Major donors
Other donors
6
4
-1
-2
1
4
6
6
GOVERNMENT 14 -3 2 13
Federal
Punjab
10
4
-2
-1
1
1
9
4
EXTERNAL INTEREST GROUPS 3 - 3 6
DOMESTIC INTEREST GROUPS 6 -1 2 7
Public-private entities
Politicians
Local partners (NGO/CSO)
2
2
2
-
-1
-
-
-
2
2
1
4
INDEPENDENT GROUPS 3 - 2 5
TOTAL 36 -7 14 43
Initially, a sample of 36 potential respondents was selected purposively from different
groups of actors in the aid policy network in Pakistan. However, during the fieldwork, the
snowballing technique allowed us to interview seven individuals in addition (see table 1).
During the three months of data collection in Pakistan – during November 2012 to January
2013 in Islamabad and Lahore – the fieldwork was successfully concluded with 43
interviewees in total.
After the collection of data from the fieldwork, the audio recordings and field notes were
transcribed, coded and analysed to elicit findings. The preliminary findings of this study
were shared with the participants to seek ‘respondent validation’. The NVivo software
program was used for handling and analysing the qualitative information. To maintain the
reliability and validity, an evaluation of the quality of this research was based largely on the
8
criterion of trustworthiness. Where appropriate, this criterion was compared and
illustrated with a number of approaches and methods, such as respondent validation,
triangulation, piloting, field records, digital recordings, and use of verbatim quotes.
Theoretical Framework
This research takes a policy network approach to explore the complex aid policy network
in Pakistan. A policy network approach is well placed to understand the complexity of
policy-making (John, 2012). A network approach deals with the complexities,
interdependencies and dynamics of public problem solving and service delivery (Klijn and
Koppenjan, 2012). This is particularly suited to the aims of this research as the policy
network approach focuses on interactions between network actors and helps understand
the policy process and public policy outcomes. Further, Overton et al. (2013) argue for a
network analysis of foreign aid that may lead us to reconceptualise aid relationships and
understand complex interactions and linkages.
Most of the policy networks literature tends to focus on Western societies (Enroth, 2011).
Moreover, the aid literature scarcely examines the role and influence of policy networks on
aid policy processes. Even less is known about how policy networks and processes operate
in Pakistan, and how multiple actors interact and bargain in game and network settings to
manage decisions. Only two known studies have examined the role of multiple actors in the
aid delivery system. Gibson et al. (2005) study complex relationships between actors in the
aid delivery system, while Bräutigam (2000) explored the choices of multiple actors within
the development aid process, and examined their interests, incentives, and the formal and
informal rules that shape possibilities for action. However, both studies warrant new
research and analysis since the context within which aid relationships are managed has
changed. Further, Hayman (2006, p. 6) concludes “the starting point of an effective agenda
for development should not be a search for global consensus, but rather a better
understanding on the ground about the constraints and perspectives of the various actors
9
from which a more country-led strategy can be devised.” This research responds to this gap
by providing renewed understanding of how the aid community operates in Pakistan.
To meet the research objective, this study uses Koppenjan & Klijn (2004) and Klijn and
Koppenjan’s (2016) actor-network analysis. The application of actor-network analysis was
considered suitable for studying the aid policy network in Pakistan because it would help
to explore the complexities, interdependencies and constraints in aid relationships in
Pakistan. This approach is novel since no other studies on foreign aid have used this
analysis as a theoretical framework. Moreover, this analysis has not been utilised to
explore policy networks in Pakistan.
THE AID POLICY NETWORK IN PAKISTAN
The aid policy network is a complex web of multiple actors involved in the aid delivery
system, connected through interdependent relations that define foreign aid policy and
priorities for the development sector. In this network, actors from different institutional
backgrounds interact to exchange information, goals and resources or means – such as
financial resources, production resources, knowledge resources, competencies and
legitimacy – to achieve their objectives.
In the case of Pakistan, this research has identified nine strategic groups of actors in the aid
policy network, see table 2. All these actors or groups of actors in the aid policy network
play roles in managing the aid policy process based on their position in the network, policy
interests and the resources they carry.
10
Table 2: Strategic groups of actors in the aid policy network in Pakistan
ACTORS/ GROUPS OF ACTORS DESCRIPTION
1. GOVERNMENT OF PAKISTAN The Pakistan government is comprised of specialised actors (or organisations). The Ministry of Finance along with its two specialised divisions, the Economic Affairs Division and the Planning Commission of Pakistan, are the key government organisations managing the aid policy process in Pakistan.
2. PROVINCIAL GOVERNMENT – PUNJAB
At the provincial level, each province has a Planning and Development department and a number of specialised provincial departments – called Executing Agency in this study – managing aid at the lower tiers. These government organisations liaise with numerous development partners, contractors and interest groups in the network to achieve governments’ objectives and policy goals.
3. MULTILATERAL DONORS This group includes multinational development organisations and finance institutions, such as the World Bank, the Asian Development Bank, the Islamic Development Bank, the UN system (consists of 18 UN agencies operating in Pakistan) and others.
4. BILATERAL DONORS – PARIS CLUB
Bilateral donors (Paris Club) focus mainly on economic policy and programing of bilateral economic assistance through international development assistance programmes, ministries or departments. Members of the Paris Club include Australia, Austria, Belgium, Canada, Denmark, France, Germany, Italy, Japan, Netherlands, New Zealand, Norway, Russia, South Korea, Spain, Sweden, Switzerland, Turkey, UK and USA.
5. BILATERAL DONORS – NON-PARIS CLUB
Bilateral donors (Non-Paris Club) focus on multiple economic areas like trade, investment, finance, banking, technical cooperation as well as economic assistance. Members of the Non-Paris Club include China, Iran, Oman, Saudi Arabia and the United Arab Emirates.
6. EXTERNAL INTEREST GROUPS This group mainly comprised of international non-government organisations (INGOs) and international research and development foundations (IRDFs) operating in Pakistan. Organisations in this group collect funds through charity or receive grants from (member) donor countries/agencies, and work in close collaboration with local partners, experts and policymakers.
7. DOMESTIC INTEREST GROUPS This group is comprised of diverse actors as compared to the others. This group includes politicians-in-power (elected politicians in the government), politicians-not-in-power (elected politicians in the opposition), other political interest groups, public-private entities, non-government organisations (NGOs), civil society organisations (CSOs), and groups or networks of local partners.
8. INTERNATIONAL FORUMS This group includes international forums – such as the United Nations, the High Level Forum on aid effectiveness of the OECD-DAC, the Commonwealth, and others – whose policies may affect aid relationships and the momentum of international development assistance in recipient countries.
9. INDEPENDENT GROUPS This group is comprised of actors from different institutional backgrounds such as researchers, civil bureaucracy, knowledge organisations (think tanks), print and electronic media organisations, and former actors (former government officials, donor officials and practitioners) as part of the aid policy network in Pakistan.
11
Actors in the Aid Policy Network
Using the actor analysis, this research determines ‘who the important actors are’ in the aid
policy network. This helps to identify actors with a leading role in determining aid policies
and development priorities; actors who are more actively involved than others in a policy
problem; and/or actors who are more influential than others based on the resources or
means available to them. Based on the mapping exercise, as part of Koppenjan and Klijn
(2004) and Klijn and Koppenjan’s (2016) actor analysis, figure 2 presents research
participants’ perceptions about the most important actors in the aid policy network in
Pakistan. The key actors (inside the central ring), important actors (inside the outer ring)
and peripheral actors (outside the outer ring) were positioned according to the
respondents’ views about actors’ possession of resources or means, policy influence,
and/or their involvement in the development process.
Figure 2: Actors in the aid policy network in Pakistan
Source: Author’s illustration
12
A majority of research participants identified the Ministry of Finance (MOF), the World
Bank, and politicians-in-power (PIP) as key actors in the aid policy network in Pakistan.
This confirms the conventional wisdom about dominant actors. These actors often lead the
policy debate and are actively involved in managing decisions related to development
priorities and the aid policy process in Pakistan.
A number of other important actors were also identified by the research participants. The
Economic Affairs Division (EAD), the Planning Commission of Pakistan (PC), USAID, the
Asian Development Bank (ADB), the International Monetary Fund (IMF), politicians-not-in-
power (PNIP), China and the civil bureaucracy were viewed as important actors in the aid
policy network in Pakistan. Participants identified these as important actors because of
their active participation in the aid policy process.
Among the bilateral donors, a considerable number of participants identified USAID as an
important and influential actor in the aid policy network in Pakistan. Statistics on foreign
aid inflows (EAD, 2002-2015) show that USAID has undoubtedly been a major bilateral
donor to Pakistan over the years. However, some of the respondents refused to accept its
importance for Pakistan. They argued that though huge grants are committed and
disbursed under US economic assistance to Pakistan, the aid is mostly off-budget, heavily
tied and conditional. Further, some participants from the recipient end and from the donor
community viewed the agency as an unreliable and unpredictable development partner
primarily due to uncertain (geo-political) bilateral relations between the US and the
Pakistan government.
Furthering the discussion, some interviewees from the government and domestic interest
groups believed that the Asian and Gulf bilateral donors are more trustworthy friends of
Pakistan than the Western (OECD) bilateral donors. Exploring respondents’ perceptions,
two reasons were identified. First, interviewees’ stances were widely based on trade and
economic sanctions Pakistan faced in the past when unlike the Western bilateral donors,
the Asian and Gulf bilateral donors remained with Pakistan. Under the US Pressler’s
Amendment, sanctions were imposed in 1990 (lifted in 1993) to restrict Pakistan
possessing a nuclear explosive device and later in response to nuclear tests in 1998 (lifted
13
in 2001). Second, participants argued that unlike the Western donors, Gulf countries
disburse on-budget grant aid that is often less conditional and tied.
Since 2008, China has been the leading official bilateral donor to Pakistan. Statistics on
foreign economic assistance to Pakistan (EAD, 2002-2015) show that China disbursed over
$2,800 million under ODA to the Pakistan government over 2010-2014. This volume of aid
is even higher than the aggregate amount received from the three leading traditional
bilateral donors to Pakistan – the US, Japan and the UK ($2,288 million altogether) – during
the same period. More recently, in April 2015, Pakistan and China signed energy and
infrastructure investment agreements worth $45 billion under the China-Pakistan
Economic Corridor (CPEC) framework. However, despite being the largest official bilateral
donor to Pakistan, the majority of respondents did not consider China as influential as
other major bilateral donors in the aid network. This was perhaps due to the limited
involvement of Chinese officials in policy forums, public policy debates at workshops and
conferences concerning MDGs, and restricted communications with other development
partners. This demonstrates the value actors attach to in-person or face-to-face
interactions in the aid policy network.
Nonetheless, the aid architecture has been transformed in Pakistan as a result of the
provincial devolution of 2011, under the 18th Constitutional Amendment. Around 17
federal ministries/divisions – previously interacting with development partners on the
MDGs – were devolved to the provincial governments. Although the transfer of power and
resources to the lower levels may improve governance, bring greater accountability, and
result in effective service delivery, the findings for Pakistan indicate that it further
complicates the system. This corresponds to extensive engagements with numerous actors,
capacity constraints at the lower tiers, and readjustments of coordination and
collaboration practices. Further, at present four different political parties have
governments in the four provinces of Pakistan. It means that each province is led by a
different political party (or alliance) with different political manifestoes and development
14
priorities.7 As Rhodes (2007) argues that devolution simply adds a further layer of
complexity that fuels territorial networks. Furthermore, restructuring governance in
Pakistan will also have an impact on the donor community, who were anticipating capacity
issues and even higher transaction costs at their end. This means that every development
partner (operating in Pakistan) will eventually have federal and provincial mission-offices
to manage and coordinate aid.
Interdependencies in Aid Relations
The dependency analysis determines the interdependencies in aid relationships exist
among different groups of network actors in Pakistan. Interdependence is commonly
constructed as mutual resource dependence, meaning that the actors in a network are
believed to be dependent on each other’s resources in order to realise objectives (Bevir,
2011; Enroth, 2011). This mutual dependence is a central assumption in the policy network
approach. However, actors with more resources and with resources which are non-
substitutable or less substitutable have more power, influence, and are less dependent on
other actors in the policy network (Gils and Klijn, 2007). Since dependence may or may not
be mutual, in a complex network setting, actors can be more or less dependent on other
actors that place them in stronger or weaker positions based on the resources available to
them. The degree that one actor is dependent upon another actor is measured by the
importance of their resource for the realisation of objectives as well as the degree to which
it is possible to acquire the resource elsewhere (Koppenjan and Klijn, 2004). Table 3
presents dependency analysis of the aid policy network in Pakistan.
The analysis suggests that major official donors and the Pakistan government are the
critical actors in the aid policy network in Pakistan. The former carries large volumes of
funds, higher quality of technical expertise, and global evidence-based knowledge
resources, while the latter carries resources such as competency, legitimacy and domestic
7 This include the Pakistan Muslim League-Nawaz (PML-N) in Punjab (and the Federal), the Pakistan Peoples Party (PPP) in Sindh, the Pakistan Tehreek e Insaaf (PTI) in Khyber Pakhtunkhwa, and the National Party & PML-N alliance in Balochistan.
15
knowledge. These resources are less substitutable in the network which leads to high
degree of interdependence and influence in strategic interactions.
Table 3: Dependency analysis
ACTORS IMPORTANT RESOURCE(S) DEGREE OF SUBSTITUTABILITY
(Low, Medium, High)
DEGREE OF DEPENDENCY (Low, Medium,
High)
CRITICAL ACTOR
(Yes/No)
MAJOR DONORS e.g., The World Bank, ADB, DFID, JICA
Financial Resource (large size aid)
Production Resources (technical expertise & TA)
Knowledge (evidence-based research & global exposure)
Low
Low
Low/Medium
High
High
High/Medium
Yes
PAKISTAN GOVERNMENT e.g., MOF, EAD, P&D Punjab, Federal and Provincial Cabinets
Competency (formal institutional authority)
Legitimacy (political support & mobilisation power)
Knowledge (social, political, and local realities, and inside information)
Low
Low
Low/Medium
High
High
High/Medium
Yes
OTHER DONORS e.g., Small bilateral donors
Financial Resource (small size aid)
Production Resources (technical expertise and capacity building)
High
High/Medium
Low
Low/Medium
No
EXTERNAL INTEREST GROUPS e.g., INGOs and IRDFs
Financial Resource (small size aid)
High Low No
DOMESTIC INTEREST GROUPS e.g., Politicians, NGO/ CSOs
Knowledge (social, political, and local realities, and community level operations)
Legitimacy (mobilisation power at local level)
High
High
Low
Low
No
FORMER ACTORS e.g., Former civil servants, experts
Knowledge (experience and exposure)
High Low No
16
It is important to mention that the degree of dependence illustrated in table 3 was assessed
from the perspective of the Pakistan government. This position helps to present a clearer
picture of interdependent donor-government aid relationships in Pakistan. For instance,
the government may express more interest in major donors because of their larger volume
of funds and technical expertise they carry as compared to other (small bilateral) donors.
Similarly, the government may be less interested in building partnership with INGOs, but
local partners (NGO/CSOs) would be more involved with INGOs because of their financial
dependence (grants). Likewise, former civil servants may not offer much to attract
governments, but sometimes they could be quite useful for development partners because
of their experience of the development process and their personal links in the government
hierarchies.
Patterns of Actors’ Interaction in the Aid Policy Network
The actor-network analysis helps to determine the frequency and diversity of interaction
patterns between major players of the aid policy network and assesses patterns of actors’
perceptions in network management. In the aid policy network some actors interact
frequently with a limited set of actors and simultaneously they interact infrequently with a
large array of players. The analysis of these patterns helps to identify central and
peripheral actors, and to make judgements about the level of inclusivity and the quality of
aid relationships in a network setting. Through interview discussions and dependency
analysis, this research finds that the Ministry of Finance, the World Bank and the
politicians-in-power occupy central places in network interactions. Figure 3 presents the
interaction patterns of these actors in the aid policy network in Pakistan.8
The thickness of bars indicates frequency of interactions that also reflects the level of
inclusivity and information transfer between network actors. The findings suggested that
the MOF stays in frequent contact with the three large multilateral donors (the World Bank,
8 The interaction patterns can be categorised as: frequent interaction (once a week); regular interaction (once in 2-4 weeks); less frequent interaction (once in 5-8 weeks); and, rare interaction (less than once every eight weeks).
17
ADB and IMF) and politicians-in-power; has regular interactions with some large bilateral
donors; and concurrently has less frequent or rare interactions with numerous
development partners and interest groups. Since aid funds are channelled to the provincial
governments through the federal account – due to the balance of payments and sovereign
guarantee requirements – the key provincial departments (P&D Board and EA) also stay in
regular contact with the federal government.
Figure 3: Patterns of actors’ interactions
Source: Author’s illustration
On the other hand, the findings also indicate that the World Bank stays in frequent contact
with the IMF, MOF and politicians-in-power. The Bank also stays in regular contact with the
Planning Commission, some leading multilateral and bilateral donors and the provincial
government departments. Further, interaction patterns of the politicians-in-power
18
indicated that they had frequent interactions with MOF and PC, and stay in regular contact
with politicians-not-in-power, officials of leading multilateral and bilateral donors, and
representatives of other political groups in the network.
Managing the Aid Policy Process in Pakistan
Perceptions are closely related to how actors define themselves, their environment and
their interests (Koppenjan and Klijn, 2004). On the basis of their perceptions, actors select
strategies and evaluate the (possible) outcomes of the aid policy process. Based on
respondents’ perceptions about managing aid policy process in Pakistan, the discussion
inferred that supply-driven foreign aid, capacity constraints in the public sector, and
governance issues were the greatest concerns for the majority of respondents within the
aid policy network.
Supply-driven foreign aid, capacity constraints in the public sector, and governance issues
were the leading problem perceptions respondents from different institutional
backgrounds had about managing the aid policy process in Pakistan. Analysis of
perceptions according to the network groups (see table 4) indicate that multilateral and
bilateral donors regarded building public sector capacity as their top priority, followed by
governance issues. The Pakistan government officials were more concerned about supply-
driven foreign aid, followed by a sub-optimal incentive structure and a lack of proper
utilisation of trained and experienced staff in the public sector. Individuals from the
domestic interest groups and former actors gave more emphasis to the need for demand-
driven foreign aid. Representatives from the external interest groups advocated greater
concentration on less developed (geographical) areas, such as South Punjab, Gilgit-
Baltistan (Northern Areas), and remote areas of Sindh and Balochistan.
19
Table 4: Patterns of actors’ perceptions about managing aid in Pakistan
ACTORS PRIME CONCERN SECONDARY CONCERN
Pakistan Government Supply-driven foreign aid Capacity constraints in the public sector: sub-optimal incentive structure; lack of proper utilisation of trained and experienced staff
Multilateral donors Capacity constraints in the public sector: policy, research, and project management skills shortage
Governance issues: low level of tax revenue collection; inefficient financial and human resource management
Bilateral donors Capacity constraints in the public sector: policy, research, and project management skills shortage
Governance issues: complex and lengthy administrative procedures; inefficient financial and human resource management
INGOs & IRDFs
(External interest groups)
Neglected less developed (geographical) areas
Governance issues: incomplete accountability and transparency
NGOs/CSOs, Politician, Public-private entities
(Domestic interest groups)
Supply-driven foreign aid Governance issues: incomplete accountability and transparency
Former actors
(Independent group)
Supply-driven foreign aid Capacity constraints in the public sector: lack of proper utilisation of trained and experienced staff
Supply-driven foreign aid
The biggest concern of the research participants, especially for the Pakistan government
officials, was that foreign aid is too supply-driven which serves to undermine the effective
value of aid in Pakistan. This finding confirms the emphasis of Altaf (2011), Bräutigam
(2000), Easterly (2006), and Riddell (2007), and Williamson (2010) who have also
indicated that supply-driven aid results in dilution of aid efforts on the ground. A number
of respondents from the government and former actors suggested that ideally foreign aid
should flow in accordance with the Pakistan government’s policies and local demands.
20
Critics of supply-driven aid in Pakistan argued that donors often suggest solutions to
domestic problems according to their own frame of reference and in the light of their own
agenda and experience, which is not always relevant to local social, economic or political
realities. In such cases, the results were not satisfactory most of the time. Respondents
indicated that sometimes instead of identifying a specific problem in the first place and
then responding according to local requirements, donors identify sectors and interventions
first, and launch their operations without giving due consideration to actual local demands.
For instance, compare allocations of foreign aid for HIV/AIDS with TB, Malaria and
Hepatitis A&B in Pakistan. A large volume of aid was allocated every year for very few
cases of HIV/AIDS in Pakistan, while insufficient amount was available for the treatment of
other diseases; resulted in thousands of deaths every year.
Among some, a perception existed that some (foreign) donor officials lack adequate
knowledge of local realities which can lead to a failure of development efforts at the local
level. Elaborating the donors’ knowledge problem, Williamson (2010, p. 11) states “Donors
are very good at specifying goals and what they hope to achieve with the aid, but they may
not know where aid is required, who it is needed by, in what locations, and in what
quantities.” Research participants indicated that foreign officials often come to Pakistan on
disbursement-oriented short-term assignments that can discourage them from
understanding local realities and digesting domestic requirements.
Though this research confirms the conventional wisdom about supply-driven foreign aid, it
is important to learn that these issues are still relevant in Pakistan today. By contrast, a
minority highlighted that the absence of an inclusive actionable development policy
framework9 and the limited capacity of the state provides opportunities for donors to
pursue their own priorities.
As of 2013, the Pakistan government lacks an inclusive actionable development policy
framework. Some interviewees indicated that a clear reflection of donors’ priorities for the
development sector can be seen in the Pakistan government’s medium to long-term 9 During 2000 and 2015, two national development plans, two medium-term development frameworks, and two long-term development visions were formulated, but only one of them – the MTDF 2005-2010 – completed its planned period; all others were either discontinued or replaced by a new policy framework.
21
development plans. Respondents argued that every new development plan is a revised
version of the previous ones which carefully articulates and acknowledges donor-led
initiatives and interventions in its policy documents so that the partnership with the donor
can be extended or established.
Capacity constraints
A large number of research participants, prominently from the donor community, indicated
capacity constraints in the public sector. Respondents identified the need to improve the
policy, research, and project management capacities of the government in a bid to make aid
efforts more productive in Pakistan. By contrast, a number of government officials argued
that the problem is more governance-oriented rather than building public sector capacity.
They indicated that expertise exists in the government, but the problem is a sub-optimal
incentive structure and a lack of proper utilisation of trained and experienced staff in the
public sector.
Governance
Some research participants identified problems and challenges related to governance in
the public sector which could undermine the state capacity to manage foreign aid. A low
level of tax revenue collection, inefficient financial and human resource management,
complex and lengthy administrative procedures, and incomplete accountability and
transparency were some of the leading governance issues identified by interviewees.
Respondents viewed these issues as hurdles for effective and collaborative development
efforts in collective choice and collective action situations, and argued for a need to
improve governance in order to increase the impact of development aid.
The research findings indicate that low revenue collection through taxation from domestic
sources was of great concern to some respondents. A few government officials advocated
concentrating more on increasing tax revenues to reduce the fiscal deficit and meet
22
domestic budgetary demands rather than borrowing from bilateral donors and multilateral
banks. Conversely, some officials concluded that it would be a difficult task to raise taxes in
the country as elites who are supposed to be taxed are either the decision-makers (such as
politicians-in-power and bureaucrats) or belong to powerful domestic interest groups
(such as politicians-not-in-power, industrialists, and landlords).
A minority, mostly outside the government, viewed complex and lengthy public
administrative procedures as a hurdle for effective and collaborative development efforts.
Respondents emphasised the need to reform these practices by simplifying procedures and
removing ‘red tape’, and bring accountability and transparency to the aid delivery system
that may help the development partners to remove process delays and lower the trust
deficit.
CONCLUSIONS
This research contributes to the aid literature and literature on policy networks by
providing rich, contextual data and an up-to-date empirical evidence of the network
structure and interactions for managing the aid policy process in Pakistan. It examines the
aid policy process in Pakistan by drawing on the policy network theory, which enables
theoretically grounded interpretations and explanations. The findings from this research
helps to gain a deeper understanding of the network structure in place, interdependencies
in aid relationships, and perceptions about constraints actors hold related to aid policy
processes in Pakistan. This qualitative approach complements the existing research in this
area that tends to use quantitative aid evaluation methods. Following are the key findings
of this research:
Carrying an important/less substitutable resource (such as large volume of funds)
does not necessarily mean an influential role within the aid policy network. For
instance, despite being the largest official bilateral donor to Pakistan, China was not
considered as influential as other major bilateral donors in the aid policy network.
This was perhaps due to the limited involvement of Chinese officials in policy
23
forums and public policy debates concerning MDGs, and restricted communications
with other development partners. Further, in strategic bilateral relations, Asian and
Gulf bilateral donors were sometimes considered to be more trustworthy
development partners than USAID and the Western (OECD-DAC) bilateral donors.
This perception was widely based on the trade and economic sanctions Pakistan
faced in the past at the hands of Western donors.
The large-scale provincial devolution of 2011 has made the aid delivery system in
Pakistan even more fragmented and complex, operating at multiple levels. These
new institutional arrangements will have an impact on managing aid policy
processes and network interactions; including extensive engagements, capacity
issues, higher transaction costs, and coordination and collaboration practices.
Last, but not least, many of the problems of international development aid are
known to the actors involved in the aid delivery system in Pakistan. However, there
has been a lack of collective action on practical steps to make foreign aid more
effective that adds to the complexity of the aid policy process. Evidence suggests
that donors tend to be pro-active and aid seems to be supply-driven in Pakistan,
partly due to the absence of an inclusive actionable development policy framework
and the limited capacity of the state. This provides opportunities for donors to
dominate the policy debate and pursue their own priorities.
Policy recommendations
In light of the research findings, this study suggests an enlightened donor approach based
on more demand-driven economic and technical assistance. A possible action to pursue an
enlightened approach could be an indigenous version of the “foundation model”.10 This
would make foreign aid less controversial by enhancing the recipients’ ownership.
However, considering unanticipated complications, the donor community could first launch
this model on a pilot basis in a specific policy sector and/or region.
10 see van de Walle, 2005
24
Further, the government – both at federal and provincial levels (and if possible, at district
level) – should work on formulating policy frameworks under their respective
development agenda. In this context, the provincial devolution of 2011 and the Sustainable
Development Goals’ (SDGs) localization approach can provide a basis for meaningful
outcome-oriented donor-government engagements; would be less reliant on the role of a
limited number of influential actors.
Finally, there is a scope for further research on policy networks and network management
in the context of international development assistance. Further research in these areas
would help policymakers, aid recipients, donors, practitioners and observers to understand
the complexity of the aid delivery process and enable them to address existing limitations.
References
ALTAF, S. W. 2011. So Much Aid, So Little Development: Stories from Pakistan. Baltimore:
The Johns Hopkins University Press.
BEVIR, M. 2011. Governance as theory, practice and dilemma. In: BEVIR, M. (ed.) The SAGE
handbook of governance. London: Sage Publications Ltd.
BEVIR, M. & RICHARDS, D. 2009. Decentring policy networks: A theoretical agenda. Public
Administration, 87 (1), pp.3-14.
BRYMAN, A. 2012. Social Research Methods. Oxford: Oxford University Press.
BRÄUTIGAM, D. 2000. Aid Dependence and Governance. Stockholm: Expert Group on
Development Issues.
BRÄUTIGAM, D. & KNACK, S. 2004. Foreign aid, institutions, and governance in Sub-
Saharan Africa. Economic Development and Cultural Change, 52 (2), pp.255-285.
BURNSIDE, C. A. & DOLLAR, D. 2000. Aid, policies and growth. The American Economic
Review, 90 (4), pp.847—868.
BURNSIDE, C. A. & DOLLAR, D. 2004. Aid, policies and growth: Revisiting the evidence.
World Bank Policy Research Working Paper, No. 3251.
EAD. 2002-2015. Status Reports on Foreign Economic Assistance. Economic Affairs Division,
Government of Pakistan: Islamabad.
EAD 2005a. Debt Management Manual. Economic Affairs Division, Government of Pakistan:
25
Islamabad.
EAD 2005b. Process Management Manual. Economic Affairs Division, Government of
Pakistan: Islamabad.
EASTERLY, W. 2006. The White Man's Burden: Why the West's efforts to aid the rest have
done so much ill and so little good. New York: Oxford University Press.
EASTERLY, W. (ed.) 2008. Reinventing Foreign Aid. Cambridge, USA: MIT Press.
ENROTH, H. 2011. Policy network theory. In: BEVIR, M. (ed.) Governance as Theory,
Practice, and Dilemma. London: Sage Publications.
FENGLER, W. & KHARAS, H. (eds.) 2010. Delivering Aid Differently: Lessons from the Field.
Washington, D.C.: Brookings Institution Press.
GIBSON, C., ANDERSSON, K., OSTROM, E. & SHIVAKUMAR, S. 2005. The Samaritan’s
Dilemma: The political economy of development aid. New York: Oxford University
Press.
GILS, M. V. & KLIJN, E-H. 2007. Complexity in decision-making: The case of the Rotterdam
harbour expansion. Planning Theory & Practice, 8 (2), pp.139-159.
GIUGALE, M. 2014. Should foreign aid be abandoned or adapted? The World Post [Online].
http://www.huffingtonpost.com/marcelo-giugale/should-foreign-aid-be-
aba_b_4555603.html [Accessed 19 July, 2016].
GOP. 2010a. Eighteenth Amendment to the Constitution of Pakistan. National Assembly of
Pakistan [Online]. http://www.na.gov.pk/en/index.php [Accessed 19 July, 2016].
GOP. 2010b. Pakistan Millennium Development Goals Report (2010). Centre for Poverty
Reduction and Social Policy Development, Planning Commission of Pakistan,
Government of Pakistan: Islamabad.
GOP. 2013. Pakistan Millennium Development Goals Report (2013). Centre for Poverty
Reduction and Social Policy Development, Planning Commission of Pakistan,
Government of Pakistan: Islamabad.
GOP. 2012-2013. Economic Survey of Pakistan [Online]. Available at: The Ministry of
Finance, Government of Pakistan http://www.finance.gov.pk/survey_1213.html
[Accessed 19 July, 2016].
GOP. 2014-2015. Economic Survey of Pakistan [Online]. Available at: The Ministry of
Finance, Government of Pakistan http://www.finance.gov.pk/survey_1415.html
[Accessed 19 July, 2016].
HAYMAN, R. 2006. The Complexity of Aid: Government strategies, donor agendas and the
coordination of development assistance in Rwanda (1994-2004). Ph.D. Edinburgh:
26
University of Edinburgh.
JOHN, P. 2012. Analyzing Public Policy. Second edition. London: Routledge
KHARAS, H. 2009. Development Assistance in the 21st Century. VIII Salamanca Forum: The
fight against hunger and poverty. Washington, D.C.: Wolfensohn Center for
Development, Brookings.
KLIJN, E.-H. 2012. New public management and governance: A comparison. In: LEVI-FAUR,
D. (ed.) The Oxford Handbook of Governance. Oxford: Oxford University Press.
KLIJN, E.-H. 2007. Managing complexity: Achieving the impossible. Critical Policy Analysis, 1
(3), pp.252-277.
KLIJN, E.-H. & KOPPENJAN, J. 2016. Governance Networks in the Public Sector. London:
Routledge.
KLIJN, E. H. & KOPPENJAN, J. 2012. Governance network theory: Past, present and future.
The Policy Press, 40 (4), pp.587-606.
KNACK, S. & RAHMAN, A. 2007. Donor fragmentation and bureaucratic quality in aid
recipients. Journal of Development Economics, 83 (1), pp.176-197.
KNACK, S. & RAHMAN, A. 2008. Donor fragmentation. In: EASTERLY, W. (ed.) Reinventing
foreign aid. Cambridge, USA: MIT Press.
KOPPENJAN, J. & KLIJN, E.-H. 2004. Managing Uncertainties in Networks. London:
Routledge.
MAVROTAS, G. (ed.) 2010. Foreign Aid and Development: Issues, challenges, and the new
agenda. UNU-WIDER: Oxford University Press.
MOYO, D. 2010. Dead Aid: Why aid is not working and how there is another way for Africa.
New York: Penguin Group.
OECD. 2015. Online data [Online]. Query Wizard for International Development Statistics
(QWIDS) http://stats.oecd.org/qwids/ [Accessed 7 February, 2017].
OVERTON, J., MURRAY, W. E. & McGREGOR. 2013. Geographies of aid : A critical research
agenda. Geography compass, 7 (2), pp. 116-127.
PBS. 2011. Pakistan Statistical Year Book [Online]. Available at: Pakistan Bureau of
Statistics http://www.pbs.gov.pk/content/pakistan-statistical-year-book-2011
[Accessed 19 July, 2016].
RHODES, R. A. W. 2007. Understanding governance: Ten years on. Organizational Studies,
28 (8), pp.1243-1264.
RIDDELL, R. C. 2007. Does Foreign Aid Really Work? New York: Oxford University Press.
27
RITCHIE, J. & LEWIS, J. 2003. Qualitative Research Practice. London: Sage.
SACHS, J. 2005a. Investing in development: A practical plan to achieve the Millennium
Development Goals. UN Millennium Project. London: United Nations Development
Programme.
SACHS, J. 2005b. The End of Poverty: Economic possibilities of our time. London: Penguin
Group.
VAN DE WALLE, N. 2005. Overcoming Stagnation in Aid-dependent Countries. Centre for
Global Development. Online: http://www.cgdev.org/publication/9781933286013-
overcoming-stagnation-aid-dependent-countries [Accessed 5 February, 2017].
WILLIAMSON, C. R. 2010. Exploring the failure of foreign aid: The role of incentives and
information. The Review of Austrian Economics, 23, pp.17-33.